A laboratory paradigm is proposed to evaluate the effects of pharmacological interventions on voluntary alcohol consumption. In this design, subjects are provided with five opportunities to complete an operant task in order to earn an alcoholic beverage or an alternative reinforcer. Should alcohol be chosen, the pace and style of drinking is monitored. In addition, assessments are made of the subject's subjective, physiological, neuropeptide, and neuroendocrine responses. The design permits examination of the rate and amount of alcohol consumed, the reinforcing value of alcohol, responses to intoxication, and the relationship between responses to alcohol and further drinking. This paradigm will be used to investigate the effects of naltrexone, a narcotic antagonist, on voluntary alcohol consumption. In clinical trials, naltrexone has been shown to result in higher rates of continuous abstinence, fewer drinking days, and lower rates of relapse than placebo. In the proposed research, the mechanisms by which naltrexone is effective will be evaluated to: 1) determine whether subjects on naltrexone exhibit greater reductions in voluntary alcohol consumption than subjects on placebo; 2) determine whether naltrexone modifies the subjective, physiological, neuropeptide, and neuroendocrine responses to alcohol; 3) evaluate whether these responses to alcohol predict further voluntary alcohol consumption; and 4) determine whether the effects of naltrexone vary according to gender. If our hypotheses regarding naltrexone's ability to reduce alcohol consumption are confirmed, this paradigm will be used to test other promising pharmacotherapies, thereby providing an important bridge between carefully controlled animal studies and full scale clinical trials in humans.